Why No Capex Recovery?

In the spring of 2012, we predicted that not only would corporate excess cash not go toward such core economic recovery “uses of funds” as CapEx, not only for the simple reason that there was, and is, no actual recovery, but that in order to create the artificial impression of improving conditions, as well as the satisfy activist investors seeking a quick ROI, companies would spend the bulk of their cash on stock buybacks and dividends. Gradually, this cash use is shifting to M&A – a classic ‘top of the cycle’ indicator – although courtesy of the unprecedented bubble in various sectors, tech most notably, corporations are opting to chiefly use overvalued stock as the currency of acquisition (see the recent purchase of Whatsapp by Facebook, funded mostly through FB stock) instead of cash. As we further explained at the same time, the main reason for this capital misallocation was simple: the Federal Reserve, whose ZIRP policy has perverted traditional hurdle rate-based capital allocation decisions, and has unleashed an all out buyback bonanza at the expense of the one cash use that is so critical to sustain not only revenue, but economic growth: capital expenditures.

As happens at the end of every year, sellside analysts and economists, all predicted that this year would be different, and the long overdue capex spending would finally be unleashed. Apparently they had far greater visibility on this matter, than on the topic of snowfall in the winter, and its disastrous impact on a $17 trillion economy, whose Q1 GDP growth forecast has cratered from 3% at the start of the year, to barely half that number currently. One of the firms that preached that the CapEx recovery is imminent is none other than Goldman Sachs, the same firm that also year after year predicts a new golden age for the US, only to see its forecast crash and burn some 4-6 months later, couched in the tried (or is that now trite) and true scapegoatings: snow, unrest in Europe, inflation or deflation in Japan, the usual. However, this time may indeed be different, and the same Goldman has just released a piece wondering “Why no capex recovery?” (despite the firm’s own forecasts to the contrary -just recall David Mericle’s “Capex: The Fundamentals Remain Strong” which now in retrospect is completely wrong).

What follows is a whole new set of “explanations” for why – once again – Goldman will have been wrong in its optimism, and why once again, we were right, after simply, and accurately, putting the blame for all that is currently wrong in the world on the one place that deserves such blame: the Federal Reserve.

Anyway, here is Goldman’s Aaron Ibbotson with Why No Capex Recovery?

Economic recovery should equal a capex recovery; that is indeed one of the key defining characteristics of the recovery phase of a business cycle. Yet we believe that “this time will be different”, certainly for developed market-based companies. Why? A combination of structural, cyclical and technological changes suggest to us that the need for capex will be lower going forward, one of the key reasons why we are cautious on capital goods.

Things are getting smaller, faster, lighter…

Ceteris paribus you need a big machine to make a big and heavy widget and a small machine to make a small and light widget; and a big machine generally demands a bigger investment than a small one. This may seem trivial, but as miniaturisation gathers pace you need less powerful motors, less space, a smaller truck to transport it around, less material to build it and much more – all with negative implications across the capex chain. In conjunction with miniaturisation machines are getting faster. A robot today can make more widgets than it could yesterday.

…until they disappear all together

The lightest and smallest widgets of them all are the ones that are now entirely virtual. As recently as the last up-cycle in 2003-2006, some companies invested capex building plants that made CDs, DVDs, video games, sat navs, maps, time tables and much more, that we now largely use our smartphones/tablets for. While capex will continue to be invested to produce our smart phones/tablets, it is difficult to envisage how it will compensate for the increasing number of goods that are becoming virtual.

Capex will be spent elsewhere…in Asia

While all capex counts, the capex we typically focus on is the that spent by listed companies in general, and listed DM companies in particular. However, the global supply chain looks very different today than it did only 10 year ago. Everything from electric components to steel is being sourced from non-DM companies, often not listed, and many of them based in China and other parts of Asia. This trend is particularly strong within tech, where Asian companies dominate the capex-intensive part of the value chain. However, myriad small Asian companies play an important part in the supply chain of many non-tech DM companies. And often, the part of the value chain that is being outsourced is the most capex-intensive part, such as producing raw materials or semiconductors, reducing both the cyclicality and the need for capex by DM-listed companies.

…and by states and private companies

The Chinese State Grid Corporation spent c.US$60 bn in 2012 and China Railway Corporation spent over US$100 bn. To put this into context, the 2,200 European non-resource companies GS covers spent in aggregate €250 bn. Over the last decade, FAI has increased by a factor of 2.5x in EMs while it has increased by only 10% in DMs. This has increased the proportion of capex spent by SOEs in a number of industries such as resources, transport and power generation and T&D. All capex counts, but this capex will not show up in the cash flow statements of the companies in our coverage, and we expect a declining slice to show up in the P&Ls of our capital goods coverage.

Too much was spent in the last up-cycle

The last cycle saw many booming end-markets: mining, power generation, shipping, O&G and Chinese construction among many growing 3x+. We do not expect this up-cycle to contain any booms, and see several of the preceding end-markets continuing (or entering) multi-year declines. At the core of our view is the long asset life of many of the capital goods sold into the booming end-markets of the last decade. This lends itself to multi-decade investment cycles. The 20-year decline in transmission and power generation capex in the US (early 1970s to late 1990s) provides a sobering example. We now believe that several end-markets are close to, or past, their peak. Of the five mentioned above, it is only O&G where we still expect growth, albeit at a substantially lower level.


Leave a Reply

Your email address will not be published. Required fields are marked *

4 + 6 =

This site uses Akismet to reduce spam. Learn how your comment data is processed.